A Long-Term Analysis of Polarization on Twitter (1703.02769v2)

Abstract: Social media has played an important role in shaping political discourse over the last decade. At the same time, it is often perceived to have increased political polarization, thanks to the scale of discussions and their public nature. In this paper, we try to answer the question of whether political polarization in the US on Twitter has increased over the last eight years. We analyze a large longitudinal Twitter dataset of 679,000 users and look at signs of polarization in their (i) network - how people follow political and media accounts, (ii) tweeting behavior - whether they retweet content from both sides, and (iii) content - how partisan the hashtags they use are. Our analysis shows that online polarization has indeed increased over the past eight years and that, depending on the measure, the relative change is 10%-20%. Our study is one of very few with such a long-term perspective, encompassing two US presidential elections and two mid-term elections, providing a rare longitudinal analysis.

An Analytical Perspective on Political Polarization on Twitter

This paper undertakes a meticulous examination of political polarization trends on Twitter over an eight-year period, an area that remains sparsely explored despite considerable interest. The authors focus on political polarization in the United States, employing a dataset of 679,000 Twitter users to assess changes in polarization within the field of social media conversations. Addressing this subject through three distinct measures—network structure, tweeting behavior, and hashtag usage—the researchers offer a comprehensive perspective on the evolution of polarization.

Objective and Methodology

The core objective is to ascertain whether political polarization on Twitter has deepened over time. The research taps into data spanning from 2009 to 2016, incorporating follower and retweet information from both political figures and media outlets with known leanings. Polarization is defined as users demonstrating a preference for either the left or right of the political spectrum, manifesting in their interactions on the platform. The analysis is segmented across three experiential dimensions:

1. Network Structure: Investigates users’ following behavior, assessing their likelihood to follow accounts from both political sides.
2. Tweet Behavior: Studies retweet patterns to understand whether users endorse content from diverse political viewpoints.
3. Content Analysis: Evaluates hashtag usage to discern the linguistic alignment and polarization among users.

Key Findings and Numerical Results

The paper reveals notable numerical results—political polarization has indeed risen by 10-20% over the eight-year period, uniformly observed across all three metrics. These significant figures are noteworthy given the diversity and volume of the dataset spanning multiple electoral cycles, including two presidential elections and two midterm elections. The augmentation in polarization is corroborated by similar increments in following and retweet behaviors, as well as hashtag utilization, building a consistent picture of polarization.

Implications and Speculations

The implications of these findings are multifaceted. Practically, an exacerbated polarized environment could have detrimental effects on the political discourse, diminishing the opportunity for dialogues across divergent ideological groups. Theoretically, these results complement existing studies indicating similar trends, such as increasing partisanship within the US Congress, thus suggesting that social media could be a reflection of offline polarization or even a vector intensifying it.

The authors also acknowledge challenges such as potential 'astroturfing' and automated tweets, though they contend that long-term account suspensions may mitigate these influences. Looking ahead, the researchers speculate that, fueled by prominent figures engaging in polarizing activities online, such as former President Trump, this trend of increasing polarization could persist unabated. These developments highlight a critical need for further longitudinal inquiries, which could be facilitated by improved access to historical social media data and more sophisticated analytical tools.

Final Observations

Overall, the paper contributes valuable insights into the longitudinal dynamics of political polarization on Twitter. It steers clear from sensationalist tones, instead bearing a clinical approach that lays a factual foundation for ongoing discourse among researchers in the domain of social media studies and political science. The results underscore the importance of continual monitoring and analysis as social media increasingly becomes embedded in the fabric of political and societal structures.